Viewing file:  xfs_behavior.h (14.58 KB) -rw-r--r--
Select action/file-type:
 (+) |  (+) |  (+) | Code (+) | Session (+) |  (+) | SDB (+) |  (+) |  (+) |  (+) |  (+) |  (+) |
/*
* Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
*
* Further, this software is distributed without any warranty that it is
* free of the rightful claim of any third person regarding infringement
* or the like. Any license provided herein, whether implied or
* otherwise, applies only to this software file. Patent licenses, if
* any, provided herein do not apply to combinations of this program with
* other software, or any other product whatsoever.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write the Free Software Foundation, Inc., 59
* Temple Place - Suite 330, Boston MA 02111-1307, USA.
*
* Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
* Mountain View, CA 94043, or:
*
* http://www.sgi.com
*
* For further information regarding this notice, see:
*
* http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
*/
#ifndef __XFS_BEHAVIOR_H__
#define __XFS_BEHAVIOR_H__
/*
* Header file used to associate behaviors with virtualized objects.
*
* A virtualized object is an internal, virtualized representation of
* OS entities such as persistent files, processes, or sockets. Examples
* of virtualized objects include vnodes, vprocs, and vsockets. Often
* a virtualized object is referred to simply as an "object."
*
* A behavior is essentially an implementation layer associated with
* an object. Multiple behaviors for an object are chained together,
* the order of chaining determining the order of invocation. Each
* behavior of a given object implements the same set of interfaces
* (e.g., the VOP interfaces).
*
* Behaviors may be dynamically inserted into an object's behavior chain,
* such that the addition is transparent to consumers that already have
* references to the object. Typically, a given behavior will be inserted
* at a particular location in the behavior chain. Insertion of new
* behaviors is synchronized with operations-in-progress (oip's) so that
* the oip's always see a consistent view of the chain.
*
* The term "interpostion" is used to refer to the act of inserting
* a behavior such that it interposes on (i.e., is inserted in front
* of) a particular other behavior. A key example of this is when a
* system implementing distributed single system image wishes to
* interpose a distribution layer (providing distributed coherency)
* in front of an object that is otherwise only accessed locally.
*
* Note that the traditional vnode/inode combination is simply a virtualized
* object that has exactly one associated behavior.
*
* Behavior synchronization is logic which is necessary under certain
* circumstances that there is no conflict between ongoing operations
* traversing the behavior chain and those dunamically modifying the
* behavior chain. Because behavior synchronization adds extra overhead
* to virtual operation invocation, we want to restrict, as much as
* we can, the requirement for this extra code, to those situations
* in which it is truly necessary.
*
* Behavior synchronization is needed whenever there's at least one class
* of object in the system for which:
* 1) multiple behaviors for a given object are supported,
* -- AND --
* 2a) insertion of a new behavior can happen dynamically at any time during
* the life of an active object,
* -- AND --
* 3a) insertion of a new behavior needs to synchronize with existing
* ops-in-progress.
* -- OR --
* 3b) multiple different behaviors can be dynamically inserted at
* any time during the life of an active object
* -- OR --
* 3c) removal of a behavior can occur at any time during the life of
* an active object.
* -- OR --
* 2b) removal of a behavior can occur at any time during the life of an
* active object
*
* For now, behavior synchornization, is controlled if CELL is
* defined.
*
* In order to allow binary compatibility with 6.5, platforms that might
* support Cellular or Cluster Irix have reserved space in 6.5 in several kernel
* structures (ex., kthread_t) which can be used to implement behavior
* synchronization functionality. Reservation of this space is controled
* by the CELL_PREPARE define.
*
* Note that currently, the CELL code, takes up more space than will be
* available in 6.5 systems. This needs to be addressed, at some point.
*
* The makefile (Makefile.kernio) that is used for compiling 3rd party
* drivers also defines CELL_PREPARE for the platforms that might
* support Cellular or Cluster Irix. In addition, this makefile also defines
* BHV_PREPARE. This causes calls to be generated to the appropriate
* BHV locking code. In 6.5, these function are stubs but they will be
* replaced with real locking code in CELL systems.
*
* Note that modifying the behavior chain due to insertion of a new behavior
* is done atomically w.r.t. ops-in-progress. This implies that even if
* CELL is off, a racing op-in-progress will always see a consistent
* view of the chain. However, correctness is not guaranteed if an
* op-in-progress is dependent on whether or not a new behavior is
* inserted while it is executing. The same applies to removal
* of an existing behavior.
*
*/
#include <linux/behavior.h>
/*
* Define stuff for behavior position masks
*/
#ifdef CELL_CAPABLE
typedef __uint64_t bhv_posmask_t;
#define BHV_POSMASK_NULL ((bhv_posmask_t) 0)
#define BHV_POSMASK_ONE(a) (((bhv_posmask_t) 1) << (a))
#define BHV_POSMASK_RANGE(a, b) (((((bhv_posmask_t) 1) << ((b)-(a)))-1) << (a))
#define BHV_POSMASK_TEST(a, b) ((a) & BHV_POSMASK(b))
#define BHV_POMASK_TESTID(a, b) BHV_POS_MASK((b)->bi_position)
#endif
/*
* Plumbing macros.
*/
#define BHV_HEAD_FIRST(bhp) (ASSERT((bhp)->bh_first), (bhp)->bh_first)
#ifdef CELL_CAPABLE
#define BHV_NEXT(bdp) (ASSERT((bdp)->bd_next), (bdp)->bd_next)
#define BHV_NEXTNULL(bdp) ((bdp)->bd_next)
#endif
#define BHV_VOBJ(bdp) (ASSERT((bdp)->bd_vobj), (bdp)->bd_vobj)
#define BHV_VOBJNULL(bdp) ((bdp)->bd_vobj)
#define BHV_PDATA(bdp) (bdp)->bd_pdata
#define BHV_OPS(bdp) (bdp)->bd_ops
#define BHV_IDENTITY(bdp) ((bhv_identity_t *)(bdp)->bd_ops)
#define BHV_POSITION(bdp) (BHV_IDENTITY(bdp)->bi_position)
// /*
// * This is used to mark an op table entry for an operation that has
// * been deleted but the entry remains reserved so that alignment
// * is maintained for compatibility for all subsequent operations.
// */
// #define BHV_OP_DELETED NULL
#ifdef CELL_CAPABLE
/*
* Macros for manipulation of behavior locks. The following
* macros operate on the lock itself. Currently, BHV locks are
* simply mrlocks but this implementation could change in the
* future. These macros should insulate us from this change.
* These macros take a mrlock_t* as an argument.
*/
#define BHV_MRACCESS(l) mraccess(l)
#define BHV_MRACCUNLOCK(l) mraccunlock(l)
#define BHV_MRTRYACCESS(l) mrtryaccess(l)
#define BHV_MRTRYPROMOTE(l) mrtrypromote(l)
#define BHV_MRUPDATE(l) mrupdate(l)
#define BHV_MRTRYUPDATE(l) mrtryupdate(l)
#define BHV_MRUNLOCK(l) mrunlock(l)
#define BHV_MRDEMOTE(l) mrdemote(l)
#define BHV_MRDIVEST(l) mrdivest(l)
#define BHV_MR_IS_READ_LOCKED(l) mrislocked_access(l)
#define BHV_MR_NOT_READ_LOCKED(l) (!mrislocked_access(l))
#define BHV_MR_IS_WRITE_LOCKED(l) mrislocked_update(l)
#define BHV_MR_NOT_WRITE_LOCKED(l) (!mrislocked_update(l))
#define BHV_MR_IS_EITHER_LOCKED(l) mrislocked_any(l)
#define BHV_MR_NOT_EITHER_LOCKED(l) (!mrislocked_any(l))
#define BHV_MR_LOCK_MINE(l) mrlock_mine(l,curthreadp)
/*
* Behavior chain lock macros - typically used by ops-in-progress to
* synchronize with behavior insertion and object migration.
* Theses macros take a behavior (bhv_head_t*) as an
* argument.
*/
#define BH_LOCK(bhp) (&(bhp)->bh_lockp->bhl_lock)
#define BHV_READ_LOCK(bhp) CELL_ONLY(BHV_MRACCESS(BH_LOCK(bhp)))
#define BHV_READ_UNLOCK(bhp) CELL_ONLY(BHV_MRACCUNLOCK(BH_LOCK(bhp)))
#define BHV_TRYACCESS(bhp) CELL_MUST(BHV_MRTRYACCESS(BH_LOCK(bhp)))
#define BHV_TRYPROMOTE(bhp) CELL_MUST(BHV_MRTRYPROMOTE(BH_LOCK(bhp)))
#define BHV_WRITE_LOCK(bhp) CELL_ONLY(BHV_MRUPDATE(BH_LOCK(bhp)))
#define BHV_WRITE_UNLOCK(bhp) CELL_ONLY(BHV_MRUNLOCK(BH_LOCK(bhp)))
#define BHV_TRYUPDATE(bhp) CELL_MUST(BHV_MRTRYUPDATE(BH_LOCK(bhp)))
#define BHV_WRITE_TO_READ(bhp) CELL_ONLY(BHV_MRDEMOTE(BH_LOCK(bhp)))
#define BHV_DEMOTE(bhp) CELL_MUST(BHV_MRDEMOTE(BH_LOCK(bhp)))
#define BHV_IS_READ_LOCKED(bhp) CELL_IF(BHV_MR_IS_READ_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_NOT_READ_LOCKED(bhp) CELL_IF(BHV_MR_NOT_READ_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_IS_WRITE_LOCKED(bhp) CELL_IF(BHV_MR_IS_WRITE_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_NOT_WRITE_LOCKED(bhp) CELL_IF(BHV_MR_NOT_WRITE_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_IS_EITHER_LOCKED(bhp) CELL_IF(BHV_MR_IS_EITHER_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_NOT_EITHER_LOCKED(bhp) CELL_IF(BHV_MR_NOT_EITHER_LOCKED(BH_LOCK(bhp)), 1)
#define BHV_LOCK_MINE(bhp) CELL_IF(BHV_MR_LOCK_MINE(BH_LOCK(bhp)), 1)
#define BHV_AM_WRITE_OWNER(bhp) \
CELL_IF(BHV_MR_IS_WRITE_LOCKED(BH_LOCK(bhp)) && \
BHV_MR_LOCK_MINE(BH_LOCK(bhp)), 1)
/*
* Request a callout to be made ((*func)(bhp, arg1, arg2, arg3, argv, argvsz))
* with the behavior chain update locked.
*
* Must have read lock before calling this routine.
* Note that the callouts will occur in the context of the last
* accessor unlocking the behavior.
*/
typedef void bhv_ucallout_t(bhv_head_t *bhp, void *, void *, caddr_t, size_t);
#define BHV_WRITE_LOCK_CALLOUT(bhp, flags, func, arg1, arg2, argv, argvsz) \
bhv_queue_ucallout(bhp, flags, func, arg1, arg2, argv, argvsz)
#define bhv_lock_init(bhp,name) CELL_ONLY(mrbhinit(BH_LOCK(bhp), (name)))
#define bhv_lock_free(bhp) CELL_ONLY(mrfree(BH_LOCK(bhp)))
#else /* not CELL_CAPABLE ie non-cell kernel */
#define BHV_READ_LOCK(bhp)
#define BHV_READ_UNLOCK(bhp)
#define BHV_NOT_READ_LOCKED(bhp) 1
#define BHV_IS_WRITE_LOCKED(bhp) 1
#define BHV_NOT_WRITE_LOCKED(bhp) 1
#endif /* CELL_CAPABLE */
#ifdef CELL_CAPABLE
extern int bhv_try_deferred_ucalloutp(bhv_head_lock_t *bhl);
static __inline int
bhv_try_deferred_ucallout(mrlock_t *mrp)
{
bhv_head_lock_t *bhl;
bhl = MR_TO_BHVL(mrp);
if (kcallout_isempty(&bhl->bhl_ucallout))
return 0;
return bhv_try_deferred_ucalloutp(bhl);
}
#endif
extern void bhv_head_init(bhv_head_t *, char *);
extern void bhv_head_destroy(bhv_head_t *);
extern void bhv_head_reinit(bhv_head_t *);
extern void bhv_insert_initial(bhv_head_t *, bhv_desc_t *);
/*
* Initialize a new behavior descriptor.
* Arguments:
* bdp - pointer to behavior descriptor
* pdata - pointer to behavior's private data
* vobj - pointer to associated virtual object
* ops - pointer to ops for this behavior
*/
#define bhv_desc_init(bdp, pdata, vobj, ops) \
{ \
(bdp)->bd_pdata = pdata; \
(bdp)->bd_vobj = vobj; \
(bdp)->bd_ops = ops; \
(bdp)->bd_next = NULL; \
}
/*
* Remove a behavior descriptor from a behavior chain.
*/
#define bhv_remove(bhp, bdp) \
{ \
if ((bhp)->bh_first == (bdp)) { \
/* \
* Remove from front of chain. \
* Atomic wrt oip's. \
*/ \
(bhp)->bh_first = (bdp)->bd_next; \
} else { \
/* remove from non-front of chain */ \
bhv_remove_not_first(bhp, bdp); \
} \
(bdp)->bd_vobj = NULL; \
}
/*
* Behavior module prototypes.
*/
#ifdef CELL_CAPABLE
extern int bhv_insert(bhv_head_t *bhp, bhv_desc_t *bdp);
extern int bhv_forced_insert(bhv_head_t *bhp, bhv_desc_t *bdp);
extern int bhv_append(bhv_head_t *bhp, bhv_desc_t *bdp);
extern int bhv_truncate(bhv_head_t *bhp, bhv_desc_t *bdp);
#endif
extern void bhv_remove_not_first(bhv_head_t *bhp, bhv_desc_t *bdp);
extern bhv_desc_t * bhv_lookup(bhv_head_t *bhp, void *ops);
extern bhv_desc_t * bhv_lookup_unlocked(bhv_head_t *bhp, void *ops);
#ifdef CELL_CAPABLE
extern bhv_desc_t * bhv_lookup_range(bhv_head_t *bhp, int lpos, int hpos);
#endif
extern bhv_desc_t * bhv_base_unlocked(bhv_head_t *bhp);
#ifdef CELL_CAPABLE
extern void bhv_global_init(void);
extern struct zone * bhv_global_zone;
extern void bhv_queue_ucallout(bhv_head_t *bhp,
int flags, bhv_ucallout_t *func,
void *, void *, caddr_t, size_t);
extern void bhv_queue_ucallout_unlocked(bhv_head_t *bhp,
int flags, bhv_ucallout_t *func,
void *, void *, caddr_t, size_t);
#endif /* CELL_CAPABLE */
/*
* Prototypes for interruptible sleep requests
* Noop on non-cell kernels.
*/
#ifdef CELL_CAPABLE
#define BLA_ACCESS 0
#define BLA_UPDATE 1
#define BLA_RWMASK 1
#define BLA_TRY 4
#define BLA_INTERRUPT 8
#ifdef BLALOG
#define bla_push(mr,rw,ra) CELL_ONLY(_bla_push(mr,rw,ra))
extern void _bla_push(mrlock_t *mrp, int rw, void *ra);
#else
#define bla_push(mr,rw,ra) CELL_ONLY(_bla_push(mr,rw))
extern void _bla_push(mrlock_t *mrp, int rw);
#endif
#define bla_pop(mrp) CELL_ONLY(_bla_pop(mrp))
extern void _bla_pop(mrlock_t *mrp);
#define bla_isleep() CELL_ONLY(_bla_isleep())
extern void _bla_isleep(void);
#define bla_iunsleep() CELL_ONLY(_bla_iunsleep())
extern void _bla_iunsleep(void);
#define bla_wait_for_mrlock(mrp) CELL_IF(_bla_wait_for_mrlock(mrp), 0)
extern uint_t _bla_wait_for_mrlock(mrlock_t *mrp);
#define bla_got_mrlock(rv) CELL_ONLY(_bla_got_mrlock(rv))
extern void _bla_got_mrlock(uint_t rv);
#define bla_curlocksheld() \
CELL_MUST((private.p_blaptr - (curthreadp)->k_blap->kb_lockp))
#define bla_klocksheld(kt) \
CELL_MUST(((kt)->k_blap->kb_lockpp - (kt)->k_blap->kb_lockp))
#endif
#endif /* __XFS_BEHAVIOR_H__ */
|